Location: Quality & Safety Assessment ResearchTitle: Rapid and early detection of salmonella serotypes with hyperspectral microscope and multivariate data analysis Author
Submitted to: Journal of Food Protection
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/12/2014
Publication Date: 3/31/2015
Citation: Eady, M.B., Park, B., Choi, S. 2015. Rapid and early detection of salmonella serotypes with hyperspectral microscope and multivariate data analysis. Journal of Food Protection. 78(4):668-674.
Interpretive Summary: Salmonella is a foodborne pathogenic bacteria and a leading cause of gastroenteritis, with some cases resulting in death. Traditional microbiological methods such as nutrient enriched plating media or polymerase chain reaction (PCR) are the standard for foodborne pathogen detection, however these methods have downsides. Traditional plating methods require several days to weeks for proper identification of the organism delaying epidemiological investigations. While PCR may be faster, there is a high cost of pathogen specific reagents in addition to an extensive training requirement. Optical detection methods offer rapid identification of bacteria based on a spectral signature unique to each organism. Hyperspectral imaging collects spectral data in a three-dimensional data cube with two dimensions representing spatial and a third relating spectral information. In this study we attempt to classify pathogenic bacterial cells based on data collected from hyperspectral microscopic images of five serotypes of Salmonella enterica; S. Enteritidis, S. Heidelberg, S. Infantis, S. Kentucky, and S. Typhimurium. Due to the similarities between spectra of five serotypes of the same species multivariate data analysis methods were utilized to discriminate between the five samples.
Technical Abstract: This study was designed to evaluate hyperspectral microscope images for early and rapid detection of Salmonella serotypes: S. Enteritidis, S. Heidelberg, S. Infantis, S. Kentucky, and S. Typhimurium at incubation times of 6, 8, 10, 12, and 24 hours. Images were collected by an acousto-optical tunable filter (AOTF) hyperspectral microscope imaging system (HMI), with a metal halide light source measuring 89 contiguous wavelengths every 4 nm between 450 – 800nm. Pearson correlation values were calculated for incubation times of 8h, 10h, and 12h, compared to 24h to evaluate the change in spectral signatures from bacterial cells over time. Regions of interests (ROIs) were analyzed at 30% of the pixels in an average cell size. Preprocessing of the spectral data was performed by applying a global data transformation algorithm, and followed with Principle Component Analysis (PCA). The Mahalanobis Distance (MD) was calculated from PCA score plots for analyzing serotype cluster separation. Partial least squares regression (PLSR) was applied for calibration and validation of the model, while Soft Independent Modeling of Class Analogy (SIMCA) was utilized to classify serotype clusters of training set. Pearson correlation values indicate very similar spectral patterns for varying incubation times ranging from 0.987 to 0.999. PCA score plots show cluster separation at all incubation times, with MD values for incubation times ranging from 2.42 to 27. PLSR had a maximum RMSEC value of 0.0025 and RMSEV value of 0.0030. SIMCA correctly classified values at 8h=98.3%, 10h=96.7%, 12h=88.3%, and 24h=98.7% with the optimal number of principle components (four or five). The results of this study suggest that Salmonella serotypes can be classified by applying PCA to HMI data as early as 8 hours incubation time.